Hand osteoarthritis phenotypes based on a biopsychosocial approach, and their associations with cross-sectional and longitudinal pain

OBJECTIVE

Hand osteoarthritis (OA) pain is characterized as heterogeneous and multifactorial. Differences in pain may be explained by underlying phenotypes, which have not been previously explored DESIGN: Latent class analysis determined classes of participants with hand OA from the Nor-Hand study baseline examination (2016-17) based on a biopsychosocial framework. Outcomes were hand and overall bodily pain intensity (Numeric Rating Scale, 0-10) at baseline and follow-up (2019-21), The relations of the classes to pain outcomes at baseline, follow-up, and change over time were analysed in separate models by linear regression, using the overall healthiest class as reference.

RESULTS

Five classes differing in radiographic hand OA burden and OA burden in the lower extremities by ultrasound, demographic factors, psychosocial burden and pain sensitization was identified. Persons with the least severe OA but higher burden of biopsychosocial factors reported the most hand pain (beta 3.65, 95% CI 2.53, 4.75). Pain was less pronounced in persons with the most severe hand OA but low burden of biopsychosocial factors (beta 1.03, 95% CI 0.41, 1.65). Results were similar for overall bodily pain and at follow-up. Changes in pain were small, but the association between a separate class defined by higher levels of biopsychosocial burden and pain changes was significant.

CONCLUSION

The five hand OA phenotypes were associated with pain at baseline and 3.5 years later. The phenotype with the least OA severity, but higher burden of biopsychosocial factors reported more pain than the phenotype with the most severe OA, reflecting the symptom-structure discordance of the hand OA pain experience.

Increased pathogenicity and pro-inflammatory capabilities of mucosal-associated invariant T cells involved in Oral Lichen Planus

BACKGROUND

Mucosal-associated invariant T (MAIT) cells assume pivotal roles in numerous autoimmune inflammatory maladies. However, scant knowledge exists regarding their involvement in the pathological progression of oral lichen planus (OLP). The focus of our study was to explore whether MAIT cells were altered across distinct clinical types of OLP.

METHODS

The frequency, phenotype, and partial functions of MAIT cells were performed by flow cytometry, using peripheral blood from 18 adults with non-erosive OLP and 22 adults with erosive OLP compared with 15 healthy adults. We also studied the changes in MAIT cells in 15 OLP patients receiving and 10 not receiving corticosteroids. Surface proteins including CD4, CD8, CD69, CD103, CD38, HLA-DR, Tim-3, Programmed Death Molecule-1 (PD-1), and related factors released by MAIT cells such as Granzyme B (GzB), interferon (IFN)-γ, tumour necrosis factor (TNF)-α, interleukin (IL)-17A, and IL-22 were detected.

RESULTS

Within non-erosive OLP patients, MAIT cells manifested an activated phenotype, evident in an elevated frequency of CD69+ CD38+ MAIT cells (p < 0.01). Conversely, erosive OLP patients displayed an activation and depletion phenotype in MAIT cells, typified by elevated CD69 (p < 0.01), CD103 (p < 0.05), and PD-1 expression (p < 0.01). Additionally, MAIT cells exhibited heightened cytokine production, encompassing GzB, IFN-γ, and IL-17A in erosive OLP patients. Notably, the proportion of CD103+ MAIT cells (p < 0.05) and GzB secretion (p < 0.01) by MAIT cells diminished, while the proportion of CD8+ MAIT cells (p < 0.05) rose in OLP patients with corticosteroid therapy.

CONCLUSIONS

MAIT cells exhibit increased pathogenicity and pro-inflammatory capabilities in OLP. Corticosteroid therapy influences the expression of certain phenotypes and functions of MAIT cells in the peripheral blood of OLP patients.

European Dairy Cattle Evaluations and International Use of Genomic Data

The European and global dairy breeding industry has benefited enormously from collaboration and sharing of data. The new era of genomics has disrupted the information flow due to the requirement to protect commercial investments. New trait phenotypes, evaluation models, and breeding goals continue to evolve and will impact the way national and proprietary data are shared and presented to the dairy industry. The global nature of cattle breeding will, however, continue to require some form of collaboration, even under the new ways of working.

Machine learning-based cluster analysis identifies four unique phenotypes of patients with degenerative cervical myelopathy with distinct clinical profiles and long-term functional and neurological outcomes

BACKGROUND

Degenerative cervical myelopathy (DCM), the predominant cause of spinal cord dysfunction among adults, exhibits diverse interrelated symptoms and significant heterogeneity in clinical presentation. This study sought to use machine learning-based clustering algorithms to identify distinct patient clinical profiles and functional trajectories following surgical intervention.

METHODS

In this study, we applied k-means and latent profile analysis (LPA) to identify patient phenotypes, using aggregated data from three major DCM trials. The combination of Nurick score, NDI (neck disability index), neck pain, as well as motor and sensory scores facilitated clustering. Goodness-of-fit indices were used to determine the optimal cluster number. ANOVA and post hoc Tukey's test assessed outcome differences, while multinomial logistic regression identified significant predictors of group membership.

FINDINGS

A total of 1047 patients with DCM (mean [SD] age: 56.80 [11.39] years, 411 [39%] females) had complete one year outcome assessment post-surgery. Latent profile analysis identified four DCM phenotypes: "severe multimodal impairment" (n = 286), "minimal impairment" (n = 116), "motor-dominant" (n = 88) and "pain-dominant" (n = 557) groups. Each phenotype exhibited a unique symptom profile and distinct functional recovery trajectories. The "severe multimodal impairment group", comprising frail elderly patients, demonstrated the worst overall outcomes at one year (SF-36 PCS mean [SD]: 40.01 [9.75]; SF-36 MCS mean [SD], 46.08 [11.50]) but experienced substantial neurological recovery post-surgery (ΔmJOA mean [SD]: 3.83 [2.98]). Applying the k-means algorithm yielded a similar four-class solution. A higher frailty score and positive smoking status predicted membership in the "severe multimodal impairment" group (OR 1.47 [95% CI 1.07-2.02] and 1.58 [95% CI 1.25-1.99, respectively]), while undergoing anterior surgery and a longer symptom duration were associated with the "pain-dominant" group (OR 2.0 [95% CI 1.06-3.80] and 3.1 [95% CI 1.38-6.89], respectively).

INTERPRETATION

Unsupervised learning on multiple clinical metrics predicted distinct patient phenotypes. Symptom clustering offers a valuable framework to identify DCM subpopulations, surpassing single patient reported outcome measures like the mJOA.

FUNDING

No funding was received for the present work. The original studies were funded by AO Spine North America.

Intricacies of aging and Down syndrome

Down syndrome is the most frequently occurring genetic condition, with a substantial escalation in risk associated with advanced maternal age. The syndrome is characterized by a diverse range of phenotypes, affecting to some extent all levels of organization, and its progeroid nature - early manifestation of aspects of the senile phenotype. Despite extensive investigations, many aspects and mechanisms of the disease remain unexplored. The current review aims to provide an overview of the main causes and manifestations of Down syndrome, while also examining the phenomenon of accelerated aging and exploring potential therapeutic strategies.

Role of Veterinary Practitioners in the Genomic Era in Dairy: Economic Impact

As veterinary practitioners serve modern dairies and ranches, genetic consultation is an area that many practitioners are implementing to bring value to their clients. As an unbiased professional with intimate knowledge of the herd's health and management practices and vision for the future, veterinarians are uniquely positioned to provide this consultative service.

The most recent progress of baicalein in its anti-neoplastic effects and mechanisms

Problems, such as toxic side effects and drug resistance of chemoradiotherapy, target therapy and immunotherapy accompanying the current anti-cancer treatments, have become bottlenecks limiting the clinical benefit for patients. Therefore, it is urgent to find promising anti-cancer strategies with higher efficacy and lesser side effects. Baicalein, a flavonoid component derived from the Chinese medicine scutellaria baicalensis, has been widely studied for its remarkable anti-cancer activity in multiple types of malignancies both at the molecular and cellular levels. Baicalein exerts its anti-tumor effects by inhibiting angiogenesis, invasion and migration, inducing cell apoptosis and cell cycle arrest, as well as regulating cell autophagy, metabolism, the tumor microenvironment and cancer stem cells with no obvious toxic side effects. The role of classic signaling pathways, such as PI3K/AKT/mTOR, MAPK, AMPK, Wnt/β-catenin, JAK/STAT3, MMP-2/-9, have been highlighted as the major targets for baicalein exerting its anti-malignant potential. Besides, baicalein can regulate the relevant non-coding RNAs, such as lncRNAs, miRNAs and circ-RNAs, to inhibit tumorigenesis and progression. In addition to the mentioned commonalities, baicalein shows some specific anti-tumor characteristics in some specific cancer types. Moreover, the preclinical studies of the combination of baicalein and chemoradiotherapy pave the way ahead for developing baicalein as an adjunct treatment with chemoradiotherapy. Our aim is to summary the role of baicalein in different types of cancer with its mechanisms based on in vitro and in vivo experiments, hoping providing proof for baicalein serving as an effective and safe compound for cancer treatment in clinic in the future.

The biochemical pattern defines MASLD phenotypes linked to distinct histology and prognosis

BACKGROUND

MASLD can manifest as hepatocellular damage, which can result in mild elevation of aminotransferases. However, in some patients, MASLD presents with cholestatic pattern.

OBJECTIVE

To assess the impact of the biochemical pattern on the natural course of MASLD, including liver damage in histology, the accuracy of non-invasive tests(NITs), and prognosis.

METHODS

Multicenter study enrolling 2156 patients with biopsy-proven MASLD, who were classified based on their[ALT/ULN)]/[(ALP/ULN)] levels at the time of biopsy: (a) hepatocellular pattern(H), > 5; (b) mixed pattern(M),2-5; (c) cholestatic pattern(C), < 2.

OUTCOMES

(a) histological evaluation of the single components of NAS, MASH, and fibrosis; (b) NITs and transient elastography assessing advanced fibrosis; (c) prognosis determined by the appearance of decompensated cirrhosis and death.

RESULTS

Out of the 2156 patients, 22.9% exhibited the H-pattern, whilst 31.7% exhibited the C-pattern. Severe steatosis, ballooning, lobular inflammation, and MASH (56.4% H vs. 41.9% M vs. 31.9% C) were more common in H-pattern (p = 0.0001),whilst C-pattern was linked to cirrhosis (5.8% H vs. 5.6% M vs. 10.9% C; p = 0.0001). FIB-4(0.74(95% CI 0.69-0.79) vs. 0.83 (95% CI 0.80-0.85); p = 0.005) and Hepamet Fibrosis Score(0.77 (95% CI 0.69-0.85) vs. 0.84 (95% CI 0.80-0.87); p = 0.044)exhibited lower AUROCs in the H-pattern. The C-pattern[HR 2.37 (95% CI 1.12-5.02); p = 0.024], along with age, diabetes, and cirrhosis were independently associated with mortality. Most patients maintained their initial biochemical pattern during the second evaluation.

CONCLUSIONS

The H-pattern exhibited greater necro-inflammation in the histology than the C-pattern, whereas the latter showed more cirrhosis. The accuracy of NITs in detecting fibrosis was decreased in H-pattern. The occurrence of decompensated events and mortality was predominant in C-pattern. Therefore, identifying MASLD phenotypes based on the biochemical presentation could be relevant for clinical practice.

Exploring phenotype-based ventilator parameter optimization to mitigate postoperative pulmonary complications: a retrospective observational cohort study

PURPOSE

To identify tidal volume (VT) and positive end-expiratory pressure (PEEP) associated with the lowest incidence and severity of postoperative pulmonary complications (PPCs) for each phenotype based on preoperative characteristics.

METHODS

The subjects of this retrospective observational cohort study were 34,910 adults who underwent surgery, using general anesthesia with mechanical ventilation. Initially, the least absolute shrinkage and selection operator regression was employed to select relevant preoperative characteristics. Then, the classification and regression tree (CART) was built to identify phenotypes. Finally, we computed the area under the receiver operating characteristic curves from logistic regressions to identify VT and PEEP associated with the lowest incidence and severity of PPCs for each phenotype.

RESULTS

CARTs classified seven phenotypes for each outcome. A probability of the development of PPCs ranged from the lowest (3.51%) to the highest (68.57%), whereas the probability of the development of the highest level of PPC severity ranged from 3.3% to 91.0%. Across all phenotypes, the VT and PEEP associated with the most desirable outcomes were within a small range of VT 7-8 ml/kg predicted body weight with PEEP of between 6 and 8 cmH2O.

CONCLUSIONS

The ranges of optimal VT and PEEP were small, regardless of the phenotypes, which had a wide range of risk profiles.

Diagnosis of cystic fibrosis: a high heterogeneity of symptoms and genotypes in a Brazil population

INTRODUCTION

In highly multiracial populations with inadequate newborn screening, knowledge of the various phenotypic presentations of Cystic Fibrosis (CF) can help reach an early diagnosis. This study aims to describe phenotypes and genotypes at the time of CF diagnosis in a state in the Northeast Region of Brazil.

METHODS

Retrospective cross-sectional study. Clinical data were extracted from the medical records of CF patients. Clinical, laboratory, and genotypic characteristics were described for patients admitted to a tertiary referral center between 2007 and 2021.

RESULTS

Fifty-eight (58) patients were included in the study, 53.5% of whom were diagnosed through clinical suspicion. The median age at diagnosis was 4.7 months (IQR: 1.5-14.8 months). Five patients had false-negative results in the newborn screening. Faltering growth was the most frequent clinical manifestation. Bronchiectasis and a history of pneumonia predominated in those older than ten, while thinness, underweight, and electrolyte imbalances were more frequent in children under two. Sequencing of the CFTR gene identified 27 genotypes, with at least one class I-III variant in all patients, and nine variants that are rare, previously undescribed, or have uncertain significance (619delA, T12991, K162Q, 3195del6, 1678del > T, 124del123bp, 3121-3113 A > T). The most frequent alleles were p.Phe508del, p.Gly542*, p.Arg334Trp, and p.Ser549Arg.

CONCLUSIONS

Malnutrition and electrolyte imbalances were the most frequent phenotypes for children < 2 years and were associated with genotypes including 2 class I-III variants. Rare and previously undescribed variants were identified. The p.Gly542*, p.Arg334Trp, and p.Ser549Arg alleles were among the most frequent variants in this population.

Clinical-based phenotypes in children with pediatric post-COVID-19 condition

BACKGROUND

Pediatric post coronavirus disease 2019 (COVID-19) condition (PPCC) is a heterogeneous syndrome, which can significantly affect the daily lives of children. This study aimed to identify clinically meaningful phenotypes in children with PPCC, to better characterize and treat this condition.

METHODS

Participants were children with physician-diagnosed PPCC, referred to the academic hospital Amsterdam UMC in the Netherlands between November 2021 and March 2023. Demographic factors and information on post-COVID symptoms, comorbidities, and impact on daily life were collected. Clinical clusters were identified using an unsupervised and unbiased approach for mixed data types.

RESULTS

Analysis of 111 patients (aged 3-18 years) revealed three distinct clusters within PPCC. Cluster 1 (n = 62, median age = 15 years) predominantly consisted of girls (74.2%). These patients suffered relatively more from exercise intolerance, dyspnea, and smell disorders. Cluster 2 (n = 33, median age = 13 years) contained patients with an even gender distribution (51.5% girls). They suffered from relatively more sleep problems, memory loss, gastrointestinal symptoms, and arthralgia. Cluster 3 (n = 16, median age = 11 years) had a higher proportion of boys (75.0%), suffered relatively more from fever, had significantly fewer symptoms (median of 5 symptoms compared to 8 and 10 for clusters 1 and 2 respectively), and experienced a lower impact on daily life.

CONCLUSIONS

This study identified three distinct clinical PPCC phenotypes, with variations in sex, age, symptom patterns, and impact on daily life. These findings highlight the need for further research to understand the potentially diverse underlying mechanisms contributing to post-COVID symptoms in children.

Obtaining patient phenotypes in SARS-CoV-2 pneumonia, and their association with clinical severity and mortality

BACKGROUND

There exists consistent empirical evidence in the literature pointing out ample heterogeneity in terms of the clinical evolution of patients with COVID-19. The identification of specific phenotypes underlying in the population might contribute towards a better understanding and characterization of the different courses of the disease. The aim of this study was to identify distinct clinical phenotypes among hospitalized patients with SARS-CoV-2 pneumonia using machine learning clustering, and to study their association with subsequent clinical outcomes as severity and mortality.

METHODS

Multicentric observational, prospective, longitudinal, cohort study conducted in four hospitals in Spain. We included adult patients admitted for in-hospital stay due to SARS-CoV-2 pneumonia. We collected a broad spectrum of variables to describe exhaustively each case: patient demographics, comorbidities, symptoms, physiological status, baseline examinations (blood analytics, arterial gas test), etc. For the development and internal validation of the clustering/phenotype models, the dataset was split into training and test sets (50% each). We proposed a sequence of machine learning stages: feature scaling, missing data imputation, reduction of data dimensionality via Kernel Principal Component Analysis (KPCA), and clustering with the k-means algorithm. The optimal cluster model parameters -including k, the number of phenotypes- were chosen automatically, by maximizing the average Silhouette score across the training set.

RESULTS

We enrolled 1548 patients, each of them characterized by 92 clinical attributes (d=109 features after variable encoding). Our clustering algorithm identified k=3 distinct phenotypes and 18 strongly informative variables: Phenotype A (788 cases [50.9% prevalence] - age ∼ 57, Charlson comorbidity ∼ 1, pneumonia CURB-65 score ∼ 0 to 1, respiratory rate at admission ∼ 18 min-1, FiO2 ∼ 21%, C-reactive protein CRP ∼ 49.5 mg/dL [median within cluster]); phenotype B (620 cases [40.0%] - age ∼ 75, Charlson ∼ 5, CURB-65 ∼ 1 to 2, respiration ∼ 20 min-1, FiO2 ∼ 21%, CRP ∼ 101.5 mg/dL); and phenotype C (140 cases [9.0%] - age ∼ 71, Charlson ∼ 4, CURB-65 ∼ 0 to 2, respiration ∼ 30 min-1, FiO2 ∼ 38%, CRP ∼ 152.3 mg/dL). Hypothesis testing provided solid statistical evidence supporting an interaction between phenotype and each clinical outcome: severity and mortality. By computing their corresponding odds ratios, a clear trend was found for higher frequencies of unfavourable evolution in phenotype C with respect to B, as well as more unfavourable in phenotype B than in A.

CONCLUSION

A compound unsupervised clustering technique (including a fully-automated optimization of its internal parameters) revealed the existence of three distinct groups of patients - phenotypes. In turn, these showed strong associations with the clinical severity in the progression of pneumonia, and with mortality.

Thyroid autoimmunity in different phenotypes of polycystic ovary syndrome: a single-center experience

PURPOSE

Polycystic ovary syndrome (PCOS) has been associated with Hashimoto's thyroiditis (HT) and 4 phenotypes have been described in this syndrome. The aim of this work was to investigate the frequency of anti-thyroid antibodies (TAb) and thyroid function in the 4 phenotypes of PCOS.

PATIENTS

This study included 448 patients with PCOS: 260 (58.0%) with phenotype A, 119 (26.6%) with phenotype B, 38 (8.5%) with phenotype C and 31 (6.9%) with phenotype D.

RESULTS

TAb positivity was detected in 90/448 patients (20.1%) and was statistically significant higher (p = 0.03) in the grouped phenotypes A-B (83/379, 21.9%) than in phenotypes C-D (7/69, 10.1%). Positive anti-thyroglobulin antibodies (TgAb) were detected in 74/448 (16.5%) patients and positive anti-thyroperoxidase antibodies (TPOAb) in 66/448 (14.7%) patients. Both TgAb and TPOAb positivity was higher but not statistically significant in phenotype A-B than phenotype C-D. High titer TgAb (> 100 UI/ml) frequency was significantly higher (p = 0.005) in grouped phenotypes A-B (39/379, 10.3%) than in phenotypes C-D (0/69, 0.0%), while no significant difference was observed for low titer TgAb (≤ 100 UI/ml). According to a binary logistic regression analysis hypothyroidism was significantly associated with TAb positivity (OR 4.19; CI 2.25-7.79; p < 0.01) but not with PCOS phenotype. Androgen profile was not associated with TAb positivity.

CONCLUSION

A higher frequency of positive TAb and of high titer TgAb and TPOAb have been detected in PCOS women with phenotypes A and B, probably in relation to the greater imbalances between estrogen and progesterone levels present in these phenotypes.

Towards personalized medicine for the treatment of equine asthma

Although horses with asthma share similar clinical signs, the heterogeneity of the disease in terms of severity, triggering factors, inflammatory profile, and pathological features has hindered our ability to define biologically distinct subgroups. The recognition of phenotypes and endotypes could enable the development of precision medicine, including personalized, targeted therapy, to benefit affected horses. While in its infancy in horses, this review outlines the phenotypes of equine asthma and discusses how knowledge gained from targeted therapy in human medicine can be applied to evaluate the potential opportunities for personalized medicine in equine asthma and to suggest avenues for research to advance this emerging field.

Clinical and imaging outcomes of different phenotypes of axial spondyloarthritis: 5-year analysis of the DESIR cohort

OBJECTIVES

To compare the long-term outcomes of three phenotypes of axial SpA (axSpA).

METHODS

Patients with a clinical diagnosis of axSpA from the DESIR cohort were grouped into three phenotypes at baseline: 'Pure axSpA' ('Axial'), 'axSpA with peripheral signs' ('IBP+Peripheral') and 'axSpA at risk' ('At risk') by latent class analysis. Clinical and imaging data were collected up to 5 years. Clinical outcomes, measured in each visit, included disability (BASFI) and quality of life (QoL; SF36). Imaging outcomes included inflammatory and structural lesions on MRI and radiographs of spine and SIJ. The association between phenotype membership at baseline and each outcome was tested in multivariable GEE models.

RESULTS

In total, 576 patients with axSpA were included. 'At risk' patients had worse disability and QoL than 'Axial' patients over time. For instance, 'At risk' patients had on average 0.4 more points in BASFI over time than 'Axial' patients [β (95 % CI): 0.4 (0.2; 0.7)]. This difference was mostly noted in female patients who were HLA-B27 positive. In addition, the difference between the 'At risk' and 'Axial' phenotypes was higher in patients receiving bDMARDs than in those not (β=0.6 vs 0.5), since BASFI improved more in 'Axial' (∆BASFI: -1.3) than in 'At risk' (∆BASFI: -0.9) treated patients. There were no differences in disability and QoL between 'Axial' and 'IBP+Peripheral' patients. Imaging outcomes were worse in the 'Axial' phenotype than in the others over time.

CONCLUSION

Patients with 'axSpA at risk' show worse self-reported outcomes over time and are less likely to benefit from anti-inflammatory treatment than those with a classical axSpA phenotype.

Personalized anti-reflux surgery: connecting GERD phenotypes in 690 patients to outcomes

BACKGROUND

Anti-reflux operations are effective treatments for GERD. Despite standardized surgical techniques, variability in post-operative outcomes persists. Most patients with GERD possess one or more characteristics that augment their disease and may affect post-operative outcomes-a GERD "phenotype". We sought to define these phenotypes and to compare their post-operative outcomes.

METHODS

We performed a retrospective review of a prospective gastroesophageal database at our institution, selecting all patients who underwent an anti-reflux procedure for GERD. Patients were grouped into different phenotypes based on the presence of four characteristics known to play a role in GERD: hiatal or paraesophageal hernia (PEH), hypotensive LES, esophageal dysmotility, delayed gastric emptying (DGE), and obesity. Patient-reported outcomes (GERD-HRQL, dysphagia, and reflux symptom index (RSI) scores) were compared across phenotypes using the Wilcoxon rank-sum test.

RESULTS

690 patients underwent an anti-reflux procedure between 2008 and 2022. Most patients underwent a Nissen fundoplication (302, 54%), followed by a Toupet or Dor fundoplication (205, 37%). Twelve distinct phenotypes emerged. Non-obese patients with normal esophageal motility, normotensive LES, no DGE, with a PEH represented the most common phenotype (134, 24%). The phenotype with the best post-operative GERD-HRQL scores at one year was defined by obesity, hypotensive LES, and PEH, while the phenotype with the worst scores was defined by obesity, ineffective motility, and PEH (1.5 ± 2.4 vs 9.8 ± 11.4, p = 0.010). There was no statistically significant difference in GERD-HRQL, dysphagia, or RSI scores between phenotypes after five years.

CONCLUSIONS

We have identified distinct phenotypes based on common GERD-associated patient characteristics. With further study these phenotypes may aid surgeons in prognosticating outcomes to individual patients considering an anti-reflux procedure.

Applicability of an unsupervised cluster model developed on first wave COVID-19 patients in second/third wave critically ill patients

OBJECTIVE

To validate the unsupervised cluster model (USCM) developed during the first pandemic wave in a cohort of critically ill patients from the second and third pandemic waves.

DESIGN

Observational, retrospective, multicentre study.

SETTING

Intensive Care Unit (ICU).

PATIENTS

Adult patients admitted with COVID-19 and respiratory failure during the second and third pandemic waves.

INTERVENTIONS

None.

MAIN VARIABLES OF INTEREST

Collected data included demographic and clinical characteristics, comorbidities, laboratory tests and ICU outcomes. To validate our original USCM, we assigned a phenotype to each patient of the validation cohort. The performance of the classification was determined by Silhouette coefficient (SC) and general linear modelling. In a post-hoc analysis we developed and validated a USCM specific to the validation set. The model's performance was measured using accuracy test and area under curve (AUC) ROC.

RESULTS

A total of 2330 patients (mean age 63 [53-82] years, 1643 (70.5%) male, median APACHE II score (12 [9-16]) and SOFA score (4 [3-6]) were included. The ICU mortality was 27.2%. The USCM classified patients into 3 clinical phenotypes: A (n = 1206 patients, 51.8%); B (n = 618 patients, 26.5%), and C (n = 506 patients, 21.7%). The characteristics of patients within each phenotype were significantly different from the original population. The SC was -0.007 and the inclusion of phenotype classification in a regression model did not improve the model performance (0.79 and 0.78 ROC for original and validation model). The post-hoc model performed better than the validation model (SC -0.08).

CONCLUSION

Models developed using machine learning techniques during the first pandemic wave cannot be applied with adequate performance to patients admitted in subsequent waves without prior validation.

Phenotypes of Mineral Bone Disorder in Chronic Kidney Disease in a Dialysis Population

BACKGROUND

Chronic Kidney Disease-Mineral and Bone Disorder (CKD-MBD) is associated with clinical outcomes. It is necessary to identify the phenotype to make clinical decisions that optimize resources and follow-up.

OBJECTIVE

To determine the frequency of the CKD-MBD phenotype in dialysis patients and the associated factors.

METHODS

Cross-sectional study in 440 patients, evaluated for CKD-MBD. Phenotypes show frequency of high, low or on target levels of PTH, vitamin D and phosphorus. The most common phenotype was used for comparisons.

RESULTS

Age was 37.5 ± 15.8 years, 53% male, 28% were diabetic, 60% on peritoneal dialysis (PD), dialysis vintage was 12.0 months (IQR 3.0-34.3). High PTH was 58%, low vitamin D 82%, high phosphorus 39%, low calcium 50%, and vascular calcification 55%. The combination of high PTH and low vitamin D and high on-target phosphorus was 39%. Those with high PTH and low vitamin D were more likely to use PD (71 vs 51%; p <0.0001), had higher lipids: total cholesterol (159 vs. 152; p = 0.002) and triglycerides (137 vs. 123; p = 0.02), higher potassium (4.7 ± 0.7 vs. 4.9 ± 0.9 mg/dL; p = 0.04), and higher serum creatinine (11.9 ± 4.4 vs. 10.6 ± 3.7 mg/dL; p = 0.01). Predictors of the most common phenotypes were PD use, total cholesterol, and serum creatinine.

CONCLUSIONS

More than one third (38%) of our sample of patients had high PTH and low vitamin D with either high or normal phosphorus. Patients with these phenotypes more frequently used PD, had higher lipids and low potassium. PD use, total cholesterol and serum creatinine were significantly associated with these phenotypes.

Identification of sleep phenotypes in COPD using machine learning-based cluster analysis

BACKGROUND

Disturbed sleep in patients with COPD impact quality of life and predict adverse outcomes.

RESEARCH QUESTION

To identify distinct phenotypic clusters of patients with COPD using objective sleep parameters and evaluate the associations between clusters and all-cause mortality to inform risk stratification.

STUDY DESIGN AND METHODS

A longitudinal observational cohort study using nationwide Veterans Health Administration data of patients with COPD investigated for sleep disorders. Sleep parameters were extracted from polysomnography physician interpretation using a validated natural language processing algorithm. We performed cluster analysis using an unsupervised machine learning algorithm (K-means) and examined the association between clusters and mortality using Cox regression analysis, adjusted for potential confounders, and visualized with Kaplan-Meier estimates.

RESULTS

Among 9992 patients with COPD and a clinically indicated baseline polysomnogram, we identified five distinct clusters based on age, comorbidity burden and sleep parameters. Overall mortality increased from 9.4 % to 42 % and short-term mortality (<5.3 years) ranged from 3.4 % to 24.3 % in Cluster 1 to 5. In Cluster 1 younger age, in 5 high comorbidity burden and in the other three clusters, total sleep time and sleep efficiency had significant associations with mortality.

INTERPRETATION

We identified five distinct clinical clusters and highlighted the significant association between total sleep time and sleep efficiency on mortality. The identified clusters highlight the importance of objective sleep parameters in determining mortality risk and phenotypic characterization in this population.

Machine learning clustering of adult spinal deformity patients identifies four prognostic phenotypes: a multicenter prospective cohort analysis with single surgeon external validation

BACKGROUND CONTEXT

Among adult spinal deformity (ASD) patients, heterogeneity in patient pathology, surgical expectations, baseline impairments, and frailty complicates comparisons in clinical outcomes and research. This study aims to qualitatively segment ASD patients using machine learning-based clustering on a large, multicenter, prospectively gathered ASD cohort.

PURPOSE

To qualitatively segment adult spinal deformity patients using machine learning-based clustering on a large, multicenter, prospectively gathered cohort.

STUDY DESIGN/SETTING

Machine learning algorithm using patients from a prospective multicenter study and a validation cohort from a retrospective single center, single surgeon cohort with complete 2-year follow up.

PATIENT SAMPLE

About 805 ASD patients; 563 patients from a prospective multicenter study and 242 from a single center to be used as a validation cohort.

OUTCOME MEASURES

To validate and extend the Ames-ISSG/ESSG classification using machine learning-based clustering analysis on a large, complex, multicenter, prospectively gathered ASD cohort.

METHODS

We analyzed a training cohort of 563 ASD patients from a prospective multicenter study and a validation cohort of 242 ASD patients from a retrospective single center/surgeon cohort with complete two-year patient-reported outcomes (PROs) and clinical/radiographic follow-up. Using k-means clustering, a machine learning algorithm, we clustered patients based on baseline PROs, Edmonton frailty, age, surgical history, and overall health. Baseline differences in clusters identified using the training cohort were assessed using Chi-Squared and ANOVA with pairwise comparisons. To evaluate the classification system's ability to discern postoperative trajectories, a second machine learning algorithm assigned the single-center/surgeon patients to the same 4 clusters, and we compared the clusters' two-year PROs and clinical outcomes.

RESULTS

K-means clustering revealed four distinct phenotypes from the multicenter training cohort based on age, frailty, and mental health: Old/Frail/Content (OFC, 27.7%), Old/Frail/Distressed (OFD, 33.2%), Old/Resilient/Content (ORC, 27.2%), and Young/Resilient/Content (YRC, 11.9%). OFC and OFD clusters had the highest frailty scores (OFC: 3.76, OFD: 4.72) and a higher proportion of patients with prior thoracolumbar fusion (OFC: 47.4%, OFD: 49.2%). ORC and YRC clusters exhibited lower frailty scores and fewest patients with prior thoracolumbar procedures (ORC: 2.10, 36.6%; YRC: 0.84, 19.4%). OFC had 69.9% of patients with global sagittal deformity and the highest T1PA (29.0), while YRC had 70.2% exhibiting coronal deformity, the highest mean coronal Cobb Angle (54.0), and the lowest T1PA (11.9). OFD and ORC had similar alignment phenotypes with intermediate values for Coronal Cobb Angle (OFD: 33.7; ORC: 40.0) and T1PA (OFD: 24.9; ORC: 24.6) between OFC (worst sagittal alignment) and YRC (worst coronal alignment). In the single surgeon validation cohort, the OFC cluster experienced the greatest increase in SRS Function scores (1.34 points, 95%CI 1.01-1.67) compared to OFD (0.5 points, 95%CI 0.245-0.755), ORC (0.7 points, 95%CI 0.415-0.985), and YRC (0.24 points, 95%CI -0.024-0.504) clusters. OFD cluster patients improved the least over 2 years. Multivariable Cox regression analysis demonstrated that the OFD cohort had significantly worse reoperation outcomes compared to other clusters (HR: 3.303, 95%CI: 1.085-8.390).

CONCLUSION

Machine-learning clustering found four different ASD patient qualitative phenotypes, defined by their age, frailty, physical functioning, and mental health upon presentation, which primarily determines their ability to improve their PROs following surgery. This reaffirms that these qualitative measures must be assessed in addition to the radiographic variables when counseling ASD patients regarding their expected surgical outcomes.

Long-term characterization of cognitive phenotypes in children with seizures over 36 months

RATIONALE

Children with new-onset epilepsies often exhibit co-morbidities including cognitive dysfunction, which adversely affects academic performance. Application of unsupervised machine learning techniques has demonstrated the presence of discrete cognitive phenotypes at or near the time of diagnosis, but there is limited knowledge of their longitudinal trajectories. Here we investigate longitudinally the presence and progression of cognitive phenotypes and academic status in youth with new-onset seizures as sibling controls.

METHODS

282 subjects (6-16 years) were recruited within 6 weeks of their first recognized seizure along with 167 unaffected siblings. Each child underwent a comprehensive neuropsychological assessment at baseline, 18 and 36 months later. Factor analysis of the neuropsychological tests revealed four underlying domains - language, processing speed, executive function, and verbal memory. Latent trajectory analysis of the mean factor scores over 36 months identified clusters with prototypical cognitive trajectories.

RESULTS

Three unique phenotypic groups with distinct cognitive trajectories over the 36-month period were identified: Resilient, Average, and Impaired phenotypes. The Resilient phenotype exhibited the highest neuropsychological factor scores and academic performance that were all similar to controls; while the Impaired phenotype showed the polar opposite with the worst performances across all test metrics. These findings remained significant and stable over 36 months. Multivariate logistic regression indicated that age of onset, EEG, neurological examination, and sociodemographic disadvantage were associated with phenotype classification.

CONCLUSIONS

This study demonstrates the presence of diverse latent cognitive trajectory phenotypes over 36 months in youth with new-onset seizures that are associated with a stable neuropsychological and academic performance longitudinally.

Clinical, Genotypic, and Imaging Characterization of the Spectrum of ABCA4 Retinopathies

PURPOSE

To investigate the clinical and genotypic differences in the spectrum of ABCA4-associated retinopathies (ABCA4Rs).

DESIGN

Observational, cross sectional case series.

PARTICIPANTS

Sixty-six patients (132 eyes) carrying biallelic ABCA4 variants.

METHODS

Patients underwent visual acuity measurement and multimodal imaging. Clinical records were reviewed for age at onset, presenting symptoms, genetic variants, and electroretinogram (ERG). Each eye was assigned to a phenotype based on age at onset, imaging and ERG: cone dystrophy-bull's-eye maculopathy (CD-BEM, 40 eyes), cone-rod dystrophy (CRD, 12 eyes), Stargardt disease (SD, 28 eyes), late-onset SD (LO-SD, 38 eyes), and fundus flavimaculatus (14 eyes). Images were analyzed for: peripapillary sparing, retinal pigment epithelium (RPE) atrophy (definitely decreased autofluorescence, DDAF), flecks patterns using autofluorescence; type of atrophy according to Classification of Atrophy Meeting reports, macular and choroidal thickness on OCT; and choriocapillaris flow deficits on OCT angiography.

MAIN OUTCOME MEASURES

Primary outcome was to report the demographic, genotypic, and imaging characteristics of the different ABCA4R phenotypes. Secondary objectives included the assessment of imaging biomarkers as outcome measures for clinical trials.

RESULTS

Age at onset was lower in CRD (12 ± 8 years) and higher in patients with LO-SD (59 ± 9 years) (all P < 0.01). Central vision loss was a common presenting symptom in CD-BEM and SD, whereas patients with LO-SD primarily complained of difficult dark adaptation. Missense variants were more frequent in CD-BEM, and splice site in CRD and LO-SD (P < 0.05). Peripapillary sparing was absent in 3 eyes with LO-SD (8%). Cone dystrophy-bull's-eye maculopathy eyes typically had complete outer retinal atrophy alterations (98%), whereas CRD and SD eyes showed both complete outer retinal atrophy and complete RPE and outer retinal atrophy (cRORA) (71%-100%). Patients with LO-SD had larger areas of DDAF (100% cRORA) and of choriocapillaris flow deficits (all P < 0.01). Repeatability of DDAF measurements was low for some phenotypes (CD-BEM and CRD) and atrophic areas <7.5 mm2. Resorbed flecks were significantly associated with CRD and LO-SD (P < 0.01).

CONCLUSIONS

This research provides a thorough evaluation of the spectrum of ABCA4R. Our findings suggest that certain phenotypes show preferential photoreceptor degeneration (e.g., CD-BEM), whereas others have substantial RPE and choriocapillaris alterations (e.g., LO-SD). We recommend that clinical trial end points take into consideration these imaging features to improve the interpretation of their results.

FINANCIAL DISCLOSURE(S)

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Identifying and understanding cognitive profiles in multiple sclerosis: a role for visuospatial memory functioning

BACKGROUND

The heterogeneous nature of cognitive impairment in people with multiple sclerosis (PwMS) hampers understanding of the underlying mechanisms and developing patient-tailored interventions. We aim to identify and classify cognitive profiles in PwMS, comparing these to cognitive status (preserved versus impaired).

METHODS

We included 1213 PwMS (72% female, age 45.4 ± 10.7 years, 83% relapsing-remitting MS). Cognitive test scores were converted to Z-scores compared to healthy controls for the functions: attention, inhibition, information processing speed (IPS), verbal fluency and verbal/visuospatial memory. Concerning cognitive status, impaired cognition (CI) was defined as performing at Z ≤ - 1.5 SD on ≥ 2 functions. Cognitive profiles were constructed using latent profile analysis on all cognitive functions. Cognitive profiles or status was classified using gradient boosting decision trees, providing the importance of each feature (demographics, clinical, cognitive and psychological functioning) for the overall classification.

RESULTS

Six profiles were identified, showing variations in overall performance and specific deficits (attention, inhibition, IPS, verbal fluency, verbal memory and visuospatial memory). Across the profiles, IPS was the most impaired function (%CI most preserved profile, Profile 1 = 22.4%; %CI most impaired profile, Profile 6 = 76.6%). Cognitive impairment varied from 11.8% in Profile 1 to 95.3% in Profile 6. Of all cognitive functions, visuospatial memory was most important in classifying profiles and IPS the least (area under the curve (AUC) = 0.910). For cognitive status, IPS was the most important classifier (AUC = 0.997).

CONCLUSIONS

This study demonstrated that cognitive heterogeneity in MS reflects a continuum of cognitive severity, distinguishable by distinct cognitive profiles, primarily explained by variations in visuospatial memory functioning.

MicroRNAs correlate with bacillary index and genes associated to cell death processes in leprosy

Mycobacterium leprae infects skin and peripheral nerves causing a broad of clinical forms. MicroRNAs (miRNAs) control immune mechanisms such as apoptosis, autophagy as well as to target genes leading to abnormal proliferation, metastasis, and invasion of cells. Herein we evaluated miRNAs expression for leprosy phenotypes in biopsies obtained from patients with and without reactions. We also correlated those miRNAs with both, bacillary index (BI) and genes involved in the micobacteria elimination process. Our results show a significant increase in the miR-125a-3p expression in paucibacillary (PB) patients vs multibacillary (MB) subjects (p = 0.007) and vs reversal reactions (RR) (p = 0.005), respectively. Likewise, there was a higher expression of miR-125a-3p in patients with erythema nodosum leprosum (ENL) vs MB without reactions (p = 0.002). Furthermore, there was a positive correlation between miR-125a-3p, miR-146b-5p and miR-132-5p expression and BI in patients with RR and ENL. These miRNAS were also correlated with genes such as ATG12 (miR-125a-3p), TNFRSF10A (miR-146b-5p), PARK2, CFLAR and STX7 (miR-132-5p). All together we underpin a role for these miRNAs in leprosy pathogenesis, implicating mechanisms such as apoptosis and autophagy in skin. The miR-125a-3p might have a distinct role associated with PB phenotype and ENL in MB patients.